Text-to-Audio Generation on Human Evaluation Subjective Audio Assessment (test)
0.2486Z-Score (OVL)TANGOFLUX
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| TANGOFLUX2024.12 | 0.2486 | 0.6919 | 1.7 | 2 | 1.1 | 1 | 1,501 | 1,628 | |
| Stable Audio Open2024.12 | 0.0723 | -0.3584 | 2.4 | — | 3.3 | 3 | 1,444 | 1,268 | |
| Tango 22024.12 | -0.019 | 0.1602 | 2.4 | 2 | 1.9 | 2 | 1,419 | 1,507 | |
| AudioLDM 22024.12 | -0.302 | -0.4936 | 3.5 | 4 | 3.7 | 4 | 1,236 | 1,196 |